Systems and Methods for Analyzing Data Element Distribution Across a Network

1. A method comprising: selecting, by a processor of a computing device, a control population of persons and a targeted population of persons, the targeted population of persons being different than the control population of persons; broadcasting, on a broadcast medium, a first plurality of media data elements associated with a product or service to the control population and a second plurality of media data elements associated with the product or service to the targeted population; receiving, at a feed repository, viewer data related to the broadcasting over a time period; loading the viewer data into an intermediate format for cleansing, adding identifiers, and extracting personal identifiable information, the personal identifiable information being routed to a separate secure storage pipeline; generating, by the computing device, a first model comprising an effect that a concentration of a media data element electronically distributed to the selected population for the product or service has on conversion metrics data by comparing the viewer data; calculating, for the media data element, a degree of targetedness measuring how close the population of persons is to a determined target of an advertiser; generating, by the computing device in real-time, a second model comprising an effect that the degree of targetedness has on the conversion metrics data; generating, by the computing device, a multi-dimensional model based on the first model and the second model that determines combined effects of the concentration of the media data element and the degree of targetedness on the conversion metrics data; generating, for a graphical user interface, a dynamic report that displays the combined effects of the concentration of the media data element and the degree of targetedness on the conversion metrics data, wherein the graphical user interface receives input for properties related to the combined effects; combining conversion estimates generated based on the multi-dimensional model with inventory quantity availability data for the product or service; predicting, by an analytics engine, future inventory quantities for the product or service based at least on the combined conversion estimates and a planned application of the concentration of media data elements; transmitting the concentration of media data elements for broadcasting based on the predicted future inventory quantities; receiving, at the graphical user interface, an input by the user that adjusts one or more of the properties related to the combined effects, the one or more properties comprising an advertising budget or advertising weight; automatically modifying in real time, by the processor, the broadcasting of the concentration of media data elements based on optimizing the broadcasting to meet the adjusted one or more properties input by the user; and generating, for the graphical user interface, an updated dynamic report that displays the updated combined effects of the automatically modified broadcasting, based on updated viewer data of the product or service.

2. The method of claim 1, further comprising determining a media impression concentration that the population has received from a media data element for the product or service over the time period, wherein the media impression concentration is based on a number of impressions delivered to the population, wherein the concentration of the media data element is the media impression concentration.

3. The method of claim 1, wherein determining the effect that the concentration of the media data element has on the conversion metrics comprises: applying a baseline level of concentration of the media data element to one or more control groups; applying elevated levels of concentration of the media data element that are higher than the baseline level to one or more treatment groups at a same degree of targetedness; and comparing first conversion metrics of the one or more control groups to second conversion metrics of the one or more treatment groups.

4. The method of claim 3, further comprising: selecting the one or more treatment groups using a first fitness function that evaluates a suitability of a group for use as a treatment group based on at least one of media data element cost associated with the group, geographic distance of the group from other treatment groups, conversions per capita for the group, difference between a national census demographic average and demographics of the group, or a degree of targetedness; and selecting the one or more control groups using a second fitness function that evaluates a suitability of a group for use as a control group based on at least one of geographic distance of the group from the one or more treatment groups, demographic disparity between the group and the one or more control groups, or a cable penetration disparity between the group and the one or more treatment groups.

5. The method of claim 3, further comprising: analyzing historical conversions for the one or more treatment groups to determine a historical variability in conversions metrics for the one or more treatment groups; and determining the elevated levels of concentration of the media data element that will cause the second conversions metrics to be outside of the historical variability.

6. The method of claim 3, further comprising: after a time period, reducing the concentration of the media data element for the one or more treatment groups to the baseline level; tracking an amount of time that it takes for the second conversion metrics to decline to levels of the first conversion metrics; calculating residual conversion associated with the elevated levels of concentration of the media data element based on the amount of time; and incorporating the residual conversion into the multi-dimensional model.

7. The method of claim 1, wherein determining the effect of the degree of targetedness comprises: selecting a first population of persons with high degree of targetedness, applying a particular concentration of the media data element to the first population, and determining first conversion metrics from the first population; selecting a second population of persons with low degree of targetedness, applying the particular concentration of the media data element to the second population, and determining second conversion metrics from the second population; and comparing the first conversion metrics to the second conversion metrics.

8. The method of claim 7, wherein the degree of targetedness is calculated as a correlation coefficient between a vector of consumer demographics for the product or service and a corresponding vector of viewer demographics for viewers of the media data element.

9. The method of claim 7, further comprising: selecting the first population of persons using a first fitness function that evaluates a suitability of a group for use as a control group based on at least one of geographic distance of the group from one or more treatment groups, matched movement of conversion metrics to the one or more treatment groups, demographic disparity between the group and the one or more treatment groups, or a cable penetration disparity between the group and the one or more treatment groups; and selecting the second population of persons using a second fitness function that evaluates a suitability of a group for use as a treatment group based on at least one of media data element cost associated with the group, geographic distance of the group from other treatment groups, population of the group, conversions per capita for the group, difference between national census demographic average and demographics of the group, or a degree of correlation between consumer demographics and viewer demographics for the group.

10. A tangible, non-transitory computer readable storage medium storing instructions that, when executed by a computing system, causes the computing system to perform operations of a method comprising: selecting, by a processor of a computing device, a control population of persons and a targeted population of persons, the targeted population of persons being different than the control population of persons; broadcasting, on a broadcast medium, a first plurality of media data elements associated with a product or service to the control population and a second plurality of media data elements associated with the product or service to the targeted population; receiving, at a feed repository, viewer data related to the broadcasting over a time period; loading the viewer data into an intermediate format for cleansing, adding identifiers, and extracting personal identifiable information, the personal identifiable information being routed to a separate secure storage pipeline; generating, by the computing device, a first model comprising an effect that a concentration of a media data element electronically distributed to the selected population for the product or service has on conversion metrics data by comparing the viewer data; calculating, for the media data element, a degree of targetedness measuring how close the population of persons is to a determined target of an advertiser; generating, by the computing device in real-time, a second model comprising an effect that the degree of targetedness has on the conversion metrics data; generating, by the computing device, a multi-dimensional model based on the first model and the second model that determines combined effects of the concentration of the media data element and the degree of targetedness on the conversion metrics data; generating, for a graphical user interface, a dynamic report that displays the combined effects of the concentration of the media data element and the degree of targetedness on the conversion metrics data, wherein the graphical user interface receives input for properties related to the combined effects; combining conversion estimates generated based on the multi-dimensional model with inventory quantity availability data for the product or service; predicting, by an analytics engine, future inventory quantities for the product or service based at least on the combined conversion estimates and a planned application of the concentration of media data elements; transmitting the concentration of media data elements for broadcasting based on the predicted future inventory quantities; receiving, at the graphical user interface, an input by the user that adjusts one or more of the properties related to the combined effects, the one or more properties comprising an advertising budget or advertising weight; automatically modifying in real time, by the processor, the broadcasting of the concentration of media data elements based on optimizing the broadcasting to meet the adjusted one or more properties input by the user; and generating, for the graphical user interface, an updated dynamic report that displays the updated combined effects of the automatically modified broadcasting, based on updated viewer data of the product or service.

11. The computer readable storage medium of claim 10, the operations further comprising determining a media impression concentration that the population has received from a media data element for the product or service over the time period, wherein the media impression concentration is based on a number of impressions delivered to the population, wherein the concentration of the media data element is the media impression concentration.

12. The computer readable storage medium of claim 10, wherein determining the effect that the concentration of the media data element has on the conversion metrics comprises: applying a baseline level of concentration of the media data element to one or more control groups; applying elevated levels of concentration of the media data element that are higher than the baseline level to one or more treatment groups; and comparing first conversion metrics of the one or more control groups to second conversion metrics of the one or more treatment groups.

13. The computer readable storage medium of claim 12, the operations further comprising: selecting the one or more treatment groups using a first fitness function that evaluates a suitability of a group for use as a treatment group based on at least one of media data element cost associated with the group, geographic distance of the group from other treatment groups, population of the group, conversions per capita for the group, difference between a national census demographic average and demographics of the group, or a degree of targetedness; and selecting the one or more control groups using a second fitness function that evaluates a suitability of a group for use as a control group based on at least one of geographic distance of the group from the one or more treatment groups, demographic disparity between the group and the one or more control groups, or a cable penetration disparity between the group and the one or more treatment groups.

14. The computer readable storage medium of claim 12, the operations further comprising: analyzing historical conversions for the one or more treatment groups to determine a historical variability in conversion metrics for the one or more treatment groups; and determining the elevated levels of concentration of the media data element that will cause the second conversion metrics to be outside of the historical variability.

15. The computer readable storage medium of claim 12, the operations further comprising: after a time period, reducing the concentration of the media data element for the one or more treatment groups to the baseline level; tracking an amount of time that it takes for the second conversion metrics to decline to levels of the first conversion metrics; and calculating residual conversions associated with the elevated levels of concentration of the media data element based on the amount of time; and incorporating the residual conversions into the multi-dimensional model.

16. The computer readable storage medium of claim 10, wherein the degree of targetedness comprises a calculated correlation coefficient between a vector of consumer demographics for the product or service and a corresponding vector of viewer demographics for viewers of the media data element.

17. The computer readable storage medium of claim 16, wherein determining the effect that the degree of correlation between the consumer demographics and the viewer demographics has on the conversion metrics comprises: selecting first media with high correlation coefficient, and concentration of the media data element to the first media; selecting second media with low correlation coefficient, and applying concentration of the media data element to the second media; and comparing first conversion metrics associated with the first media to second conversion metrics associated with the second media.

18. The computer readable storage medium of claim 17, the operations further comprising: selecting the first media using a first fitness function that evaluates a suitability of media to use in a control group based on at least one of geographic distance of the group from one or more treatment groups, matched movement of conversion metrics to the one or more treatment groups, demographic disparity between the group and the one or more treatment groups, or a cable penetration disparity between the group and the one or more treatment groups; and selecting the second media using a second fitness function that evaluates a suitability of a media to use in a treatment group based on at least one of media data element cost associated with the group, geographic distance of the group from other treatment groups, population of the group, conversions per capita for the group, difference between national census demographic average and demographics of the group, or a degree of correlation between the consumer demographics and the viewer demographics for the group.

19. A system comprising a data storage device storing instructions for a method in an electronic storage medium and a processor of a computing device configured to execute the instructions to perform a method including: selecting, by the processor of the computing device, a control population of persons and a targeted population of persons, the targeted population of persons being different than the control population of persons; broadcasting, on a broadcast medium, a first plurality of media data elements associated with a product or service to the control population and a second plurality of media data elements associated with the product or service to the targeted population; receiving, at a feed repository, viewer data related to the broadcasting over a time period; loading the viewer data into an intermediate format for cleansing, adding identifiers, and extracting personal identifiable information, the personal identifiable information being routed to a separate secure storage pipeline; generating, by the computing device, a first model comprising an effect that a concentration of a media data element electronically distributed to the selected population for the product or service has on conversion metrics data by comparing the viewer data; calculating, for the media data element, a degree of targetedness measuring how close the population of persons is to a determined target of an advertiser; generating, by the computing device in real-time, a second model comprising an effect that the degree of targetedness has on the conversion metrics data; generating, by the computing device, a multi-dimensional model based on the first model and the second model that determines combined effects of the concentration of the media data element and the degree of targetedness on the conversion metrics data; generating, for a graphical user interface, a dynamic report that displays the combined effects of the concentration of the media data element and the degree of targetedness on the conversion metrics data, wherein the graphical user interface receives input for properties related to the combined effects; combining conversion estimates generated based on the multi-dimensional model with inventory quantity availability data for the product or service; predicting, by an analytics engine, future inventory quantities for the product or service based at least on the combined conversion estimates and a planned application of the concentration of media data elements; transmitting the concentration of media data elements for broadcasting based on the predicted future inventory quantities; receiving, at the graphical user interface, an input by the user that adjusts one or more of the properties related to the combined effects, the one or more properties comprising an advertising budget or advertising weight; automatically modifying in real time, by the processor, the broadcasting of the concentration of media data elements based on optimizing the broadcasting to meet the adjusted one or more properties input by the user; and generating, for the graphical user interface, an updated dynamic report that displays the updated combined effects of the automatically modified broadcasting, based on updated viewer data of the product or service.

20. The system of claim 19, wherein the multi-dimensional model is a two-dimensional model.